I just saw 5dmk3 and Nikon D800 samples on their official sites and I'm...

Started Mar 6, 2012 | Discussions thread
noirdesir Forum Pro • Posts: 13,460
Re: I just saw 5dmk3 and Nikon D800 samples on their official sites and I'm...

Skip M wrote:

But it's 2 stops behind the 5D mkIII in ISO capability.

It is two stop behind in in-camera processing capabilities , ie, if you need a jpeg from the camera at ISO of 50 or 100k, only then is the 5D III ahead of the D800 in low light capabilities. If you process the raws yourself, there is no limit to what 'ISO' you can process it to. All these really high ISO settings are anyway just a digital scaling which a raw processor can do exactly the same way as the camera itself.

And, frankly, ISO capability is more important. Also, if the D800's sensor behaves like that of the D7000, you lose all of your DR addvantage by ISO400, according to DxO.

The D800 does not behave like the D7000, it behaves like 2.25x D7000 images stitched together. You have to scale the values of the D7000 to the larger sensor area to know how the D800 performs. DxO explains here ( http://www.dxomark.com/index.php/Publications/DxOMark-Insights/Detailed-computation-of-DxOMark-Sensor-normalization ) how to scale the results. Assume you have the 'screen' results for the D800, to normalize them to DxO's standard of 8 MP, you add log2(36/8)^.5 = 1.08 to it (for DR and TR). If you had the screen results for the D800 DX crop, you'd add log2(15/8)^.5 = 0.45 to it. Thus to convert between the D800 DX crop and the fullframe results, you add look at the difference, ie, 1.08 - 0.45 = 0.63 stops of DR.

This makes the extrapolated D7000 and 5DII pretty close in DR over most of the ISO range (notwithstanding any improvements by the D800 and 5DIII). But that view is missing the a large part of the picture because you can use the D800 at a lower ISO setting and push things during raw conversion because it has a very low shadow noise at base ISO. Thus, the DR values of DxO for above base ISO values apply only when using the camera suboptimally for high DR applications under restricted light. In short, on all semi-recent Canon cameras, you have throw away highlight data (at the A/D converter) to get low shadow noise, with the recent Sony Exmor sensors, you gain little shadow noise improvements when throwing away highlight data in the A/D converter, thus you don't have to do it to get similar shadow noise as with Canon's sensors.

Of course, the performance of the 5DIII sensor is only partially known yet, so a precise comparison has to wait until we have more data.

Post (hide subjects) Posted by
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow